National Capitol Region

HAZUS User Group Call

August 21, 2014 10:00 AM EST

Listen to the recording here to follow along with the presentation:

http://www.freeconferencecalling.com/Recordings/Recording.aspx?fileid=L

AF3494_08212014070219220_1180281&bridge=697620&email=&account

id=1116753

NCR HUG Call Details

Conference Call Details: 1. Dial-in: 1-267-507-0240

2. Conference code: 697620

Join the Meeting through Adobe Connect:

(No Registration Required)

Click here to access the Adobe Connect meeting

The call will be recorded and the audio will be combined with the presentation

and sent out at the end of the call. The audio and presentation will also be

made available on the following websites:

NCR HUG Use HAZUS page - http://www.usehazus.com/ncrhug

NCR HUG LinkedIn page - http://www.linkedin.com/groups/National-Capitol-

Region-HAZUS-User-4790251?trk=myg_ugrp_ovr

2

Agenda • Welcome + Announcements

• FEMA Region III HAZUS Training - E172 HAZUS for Flood

• Statewide Geospatial Coordination – Progress Report

• Shane Hubbard – Tracking spatiotemporal patterns of buildingvulnerabilities and evacuations during flood hazards

• Cynthia McCoy – Report : HAZUS modernization and futureenhancements

• Reminder - HAZUS Data & Training Collaboration Sharepoint

• Request for Volunteers – Next NCR HUG Call

• Adjourn

3

Announcements HAZUS Conference 2014 – Monday August 4 – 6, 2014

All conference presentations can be found on the NCR HUG Page:

http://www.usehazus.com/ncrhug

4

Announcements Upcoming HAZUS Courses at EMI

The Emergency Management Institute (EMI) has the following HAZUS training course available:

E170 Hazus-MH for Hurricane: October 6-9, 2014

E190 ArcGIS for Emergency Managers: October 20-23, 2014

E176 Hazus-MH for Floodplain Managers: November 10-3, 2014

E172 Hazus-MH for Flood: December 1-4, 2014

E313 Basic HAZUS-MH: December 15-18, 2014

E317 Comprehensive Data Management for HAZUS: January 5-8, 2015

E174 Hazus-MH for Earthquake: March 30-April 2, 2015

Please note that enrollment is closed for each course, 6 weeks prior to the course!

• To apply for a HAZUS training course, please visit:

http://training.fema.gov/Apply/

• To enroll, download the Admission Application or contact the

HAZUS Training Specialist at EMI:

Jordan Manos – Jordan.manos@fema.dhs.gov

For further information on registration, please visit training.fema.gov/emiweb

5

Announcements FEMA Region III E172 HAZUS for Flood Training

This is not the same course offered in October 2013 - Focus on the State’s specific needs related to flood risk assessment.

The initial course provided the basics of HAZUS, however further training is necessary to accurately operate the HAZUS models. This course will train participants to completely operate the HAZUS flood model, understand the necessary data inputs, the quality of data inputs necessary, and how to interpret the results for decision making.

Intended for FEMA Region III Federal State, County, Local Gov. & University partners

4 day Course – Onsite session

Day 1-3: Official E172 HAZUS for Flood training course

Day 4 : Interactive session with dynamic discussion, highlighting the intersections between HAZUS and Mitigation Planning. We will discuss the State’s specific challenges related to flooding, analyzing that risk and informing mitigation strategies. We will discuss the State’s specific challenges related to flooding, analyzing that risk and informing mitigation strategies.

Both the decision maker and the GIS analyst should plan to attend as a team, to promote more valuable dialogue and awareness.

Training Sites:

Completed- Lancaster, Pennsylvania - June 16-19, 2014

Completed - Morgantown, West Virginia - July 28-31, 2014

NEXT WEEK! Dover, Delaware - August 25-28, 2014 (4 spots available!)

Trainers: Kevin Mickey and David Coats from the POLIS Center at Indiana University – Purdue University

***FEMA Region III is providing the training at no cost. However, any additional costs associated with travel will not be paid for by FEMA.

6

Statewide Geospatial Coordination: Monthly Progress Report

Current Effort

• Performing State/District level data gap analysis

• Performing Local Outreach to Counties

Next steps

• Develop outreach strategy to the Local Governments

• Outreach to the Counties using the revised list (remaining datasets).

• Develop tools and workflows – streamline input of multi-jurisdictional data

State and District update

District of Columbia – Completed District wide assessment. Final assessment can be found

here: http://www.usehazus.com/ncrhug/projects

Delaware – Completed Statewide and Local assessment. Current assessment can be found

here: http://www.usehazus.com/ncrhug/projects . Datasets will be made available on the eGIS

system in September, and this assessment will be updated with the links to these datasets.

• Maryland – MDE, MEMA and DoIT are performing their State level gap analysis

• Pennsylvania – State level gap analysis Completed. California University of PA – Local level outreach to Counties continues. Surveys due September 30th!

• Virginia – VDEM Completed State level gap analysis. Next step is to coordinate the local level

outreach.

• West Virginia – WVDHS, WVU and WV DoE Completed State level gap analysis.

Coordinating with PDCs as the first round of local level outreach.

7

Tracking spatiotemporal patterns of

building vulnerabilities and

evacuations during flood hazards

Shane HubbardJunchuan Fan

Kathleen StewartEmily White

The University of Wisconsin – Madison, SSEC/CIMSSAugust 6th, 2014

Background

June 2008, Iowa City

Record flood by 3 feet (1993)

25 buildings damaged on the UIowa Campus

Flood damage and rebuilding costs $743M – $1B (2011 Globe Gazette) – contents in the $100Ms

Classes cancelled and major research programs were delayed or even lost

Arts Campus

A Unique Evacuation

Open Research Questions

Approached by University group responsible for flood planning and response.

What are the factors that affect campus evacuations?

Critical infrastructure?

Research?

Student Housing?

How to design a model that can be used to support decision-making about building evacuations on campus?

What are the underlying spatiotemporal properties of a model for campus building evacuations?

A foundation for evacuation planning

The model will

provide a means to analyze and visualize a time-varying flood

boundary and how it intersects with different parts of the campus

show which parts of the campus are vulnerable to flooding as the

flood boundary changes over time

form the basis for prioritizing evacuations on the UI campus

based on information about the spatiotemporal properties of the

flood boundary combined with data about building location,

contents and access

Data collection: flood boundary data

Obtained a set of flood depth grids for the Iowa River from The Iowa Flood Center

Our model begins with flood depth grids that equate a 17-foot (7,180 cfs) flood stage along the University of Iowa campus

The National Weather Service (NWS) reports the University of Iowa Campus sees building damage at approx 27 feet (25,100 cfs)

Flood depth grids continue every 0.5 feet of flood stage from 17 through34 feet (55,000 cfs)

Automation

The process of creating flood

layers and formatting each into

a GIS layer viewable on a

webpage.

Automation allows real-time

depth grid development on the

fly

Coffer dam

Sandbagging efforts

Data collection: spatial data

Acquired spatial data from Facilities

Management

Updated GIS data of all campus

roads, buildings and residence

halls, sidewalks, parking lots, and

steam tunnels as provided by

Facilities Management

Statewide LiDAR data

Elevation data for campus

terrain, campus buildings, river

Campus-wide orthophotos collected

June 15th, 2008 for validation

University of Iowa FERP

The University of Iowa has a detailed FERP that has been updated

to July 2009 and more recently in 2012

Details critical buildings – building watch list

Evacuation plans

The FERP identifies 30 buildings as at-risk facilities potentially

within the flooding boundary

As part of our research, we have revealed a further seven facilities

also inside the flood boundary

Watch list buildings

Art Building West

Iowa Advanced

Technology Lab (IATL)

Mayflower

Softball Complex

Equipment Building

Theatre Building

Voxman Music Building

Hancher Auditorium*

English-Philosophy Bldg

Adler Journalism

Becker Communications

Studies Bldg

Library

Water Plant

Power Plant

Cambus Maintenance

Facility

C. Maxwell Stanley

Hydraulics Lab

Art Building

Museum of Art

Butler Storage and Hydraulics

Annexes

Hydraulics Lab Model Annex

Hydraulics Research East

Annex

Hydraulics Wind Tunnel

Annex

North Hall

Iowa Memorial Union

Danforth Chapel

Madison Street Services

Building

Fleet Service Building

Hawkeye Court Apartments

Lindquist Center

Additional buildings at possible risk

Liberty Square

P. Sue Beckwith, M.D.

Boathouse

Campus Recreation and

Wellness Center

South Capitol Services

Building

Morgan Sanitary Eng

Lab

Gilbert Street Building

Mossman Business

Services Building

Building the evacuation model

Two step process:

A spatiotemporal analysis between flood boundaries and depth gridswith University of Iowa infrastructure have been computed to determine campus areas at risk for flooding

i.e., the buildings, roads, parking lots, etc.

Based on these intersections and building elevations, we can determine what features are vulnerable with each new change in flood stage

Time is being incorporated in the model so each flood stage and associated depth grid is associated with a likely time of building inundation.

Evacuation start time =

Time of risk** – Time needed to evacuate building contents

Building Risk Layers

Flood depth grids were translated into flood elevations

Flooded buildings were identified for each flood elevation and

placed into categories

Vulnerable basement (0.5 ft grid before flooded basement)

Flooded basement (revised with 2008 Flood data)

Vulnerable first floor (0.5 ft grid before flooded basement)

Flooded first floor (flood elevation and first floor elevation)

Vulnerabilities are calculated to account for uncertainty

Data collection and preparation

Basement flooding analysis

Using data from the FERP, fields for the elevations of basements, and sub-basements of buildings on the watch-list were added to the GIS layer of university buildings

For buildings with multiple basement elevations, there is a record for each of the basement elevations

The estimated water table depth, the level of water in the basement, and the distance from the building to the nearest flooded area, were recorded for each flood depth

But during validation found this did not produce accurate results…

Used the data collected from interviews on when the basements flooded and what the corresponding flood elevation was at the time.

Data collection and preparation

First-floor flooding analysis

Data from the FERP provided first-floor building elevations. The building data layer was intersected with the flood depth elevation rasters.

When such an intersection exists, the flood elevation depth is compared to the first floor elevation for the building

If the first floor elevation is less than the flood elevation, the first floor is considered flooded

A zonal statistic is used to calculate minimum, maximum and mean water levels on the first floor. Max was used.

Building Risk Layers

Building risk is tied to a specific

flood level.

In real-time risk layers are

displayed based on the current

and forecasted flood level

Risk Analysis Results

Two peaks

50 Year

100 Year

Related to

floodplain

construction

regulations

Hazus-MH Depth/Damage Analysis

The possible damage severity to each building was tracked using

the average flood depth from a zonal statistics analysis in ArcGIS

and then assigned a damage% from the damage curves in Hazus-

MH.

There are major differences in the damage functions and the default

value does not reflect the historical depth damage relationship on

the campus.

Building content information

The direct flood risk to the building is not the entire problem….

During the summer, we interviewed representatives from each of the 30+ buildings on the watch-list

We collected data on

estimated time needed for evacuation today

time needed for evacuation during 2008 flooding

details on where contents need to be moved

challenges encountered during 2008 flooding

length of time building was closed due to flooding

key contents that would need to be evacuated

details on numbers of building residents, vehicles, etc.

Diversity of vulnerable buildings

The vulnerable buildings represent a very diverse set of needs and

times for evacuation

For example

The University Library

Mossman Business Services Building (campus mail…)

UIHC Patient Transportation Services

South Gilbert Street Building (Hawkshop, bookstore storage)

Lindquist Center – UITV, ITS

Power Plant, Water Plant, IMU…

Interview Findings

Certain research projects can be extremely sensitive to evacuations

Million dollar projects cannot be moved overnight

Residence Halls are difficult to evacuate

If summer, key faculty or administrators who would normally lead an

evacuation may be away (what do we take?)

Time available for evacuation is important…2 days? 5 days?

What is the maximum time available for evacuations?

Road and Bridge Closures

An important variable for modeling building evacuations is road/bridge closure or access to a building.

Evacuations are road closure dependent

Road and bridge closures are computed for each flood elevation

Produces 2 additional evacuations.

Evacuation Time

The recommended time evacuate is calculated by…

The time estimated to evacuate the building (determined by building

interviews)

The expected time of arrival of flooding or the loss of building

access from inundated roads or closed bridges.

Subtract the necessary time to evacuate the building from the

expected time of arrival of flooding

Conservative estimate…expected time of arrival of flooding is based

upon the vulnerable basement inundation time.

Evacuation Layers

Restoration Time and Priority

An important consideration for evacuation priority is the length of time required to make the building functional again.

Zonal statistics were used in ArcGIS to calculate the average flood depth.

The Hazus-MH restoration functions were modified to estimate the time for building restoration

University administrators may elect

to change the evacuation

recommendation time based-

upon the level of importance of

the building vs the time to

restoration

Spatiotemporal Evacuation Model

We have developed a real-time, visual, interactive model that uses a map-based interface to allow users to interact and query the spatiotemporal characteristics of a flood event and plan for possible evacuations

Google Maps API to present the model in a web-based platform.

Connected to a SQL database for data storage.

Putting everything together

Interactive model is based on

Our modeled assessments of

infrastructure at risk

Building content information

and FERP data

Time needed for evacuations

Current stream conditions and

forecasts from the NWS (or

other sources)

HawkEvac

Historical, Real-

time and

Forecasted data

Detailed Flood

information

Layer Selection

Reporting

Future evacuations based on time to

evacuate – time of flooding

6 groups of evacuations

Evacuated

Evacuations in Progress

Evacuations Beginning Today

Evacuations Beginning Tomorrow

Evacuations Beginning in 2 days

Evacuations Beginning in 3 – 7

days

Information on basements and first

floors are included

2008 Iowa City Flood Results

There is a spatiotemporal progression of flood evacuation start times across the University of Iowa campus.

Not related to any flood surge movement downstream.

Essential in understanding this pattern during real-time evacuations for congestion around evacuated buildings.

Differences for every flood.

User Interaction

Flood officials at the University requested a means for identifying what actually happens versus what we suggest should happen.

Form was designed to provide the user with designating the current evacuation status for each vulnerable building

Flash flooding

University decision

An additional layer is also present in the application.

Temporary Mitigation on Campus

Future floods on campus will be mitigated with the use of HESCO barriers.

The Iowa Flood Center used a HESCO layer to simulate the changes to the floodplain as a result of the mitigation measures.

At this time evacuation recommendations are only calculated on the NO mitigation layers.

….And model validation

Future Work

Extend model to capture recovery times

The depth of water predicted inside the

building.

Important for faculty, staff, and students.

Enhanced evacuation times based on…

Floor area that will be evacuated.

What is being evacuated.

Distance the contents must travel.

Basement flooding

Thank You

Shane Hubbard, Ph.D.

Space Science and Engineering

Center

University of Wisconsin - Madison

1225 W. Dayton St.

Madison, WI 53706

Office: (608) 263-3712

Cell: (608) 215-6059

e-mail:

shane.hubbard@ssec.wisc.edu

Hazus Modernization: Building Blocks

Stakeholder Feedback Led Modernization Approach:

There are four major focus areas for the modernization approach:

1. Technology Refresh – improved reliability, speed,and capability to take on new functionality

2. Better Alignment with FIMA Objectives

3. Improved Accuracy

4. Hazus-lite Capabilities (allowing non-technical usersto have an easy to use customizable interface)

Project Timeline: Phase 1 Hazus Modernization

August 2015August 2014 January 2015

Task 2: Functional enhancements to the Flood module,

Dasymetric GBS

Task 1: Upgrade Hazus to be compatible

with ArcGIS 10.2 and Windows 8

Task 3: Re-architecture and source code update

Task 4a: Develop Proof of Concepts, integration of NFHL

and NFIP Rating Approach

Task 4b

June 2014

11/24/14

03/24/15

08/21/15

January 2015

April 2015

Modernization Phase 1: Project Tasks (Tues Aug 5th)

Task 1: Upgrade Hazus to be compatible with ArcGIS 10.2 and Windows 8 = Hazus2.2 first major release in 2.5 years

Task 2: Functional enhancements to the Flood module (inventory and SP release)

•Apply the dasymetric GBS exposure distribution approach

•Synchronization of depth damage functions

•Implement study region aggregation at custom jurisdiction or neighborhoodlevels

Modernization Phase 1: Project Tasks (Wed Aug 6th)

Task 3: Re-architecture and source code update

•Replace MS Access (.mdb) personal Geo-Databases (pGDB) for all modules

•Migration of DTS packages to SQL Server Integration Services (SSIS)

•Replace VB6 and COM architecture with .NET (C#) for flood module

Task 4: Develop Proof of Concepts

•Feasibility of NFHL data integration with Hazus Level 1 hazard determination for 1% and0.2% flood risk

•Performing building and economic AAL using a combination of NFHL and Hazus floodreturn periods

Modernization Vision:

Improved Accuracy: Enhanced site-specific capability. Enhance Level 1 losses improving block level losses by implementing dasymetric methodology

Hazus-lite Capabilities: Provide Hazus capabilities to a larger non-technical audience through a web-based interface.

Modernization Vision:

Web-Enabling: Many good examples (NC FRIS, ImageCat Inlet, RMS One, etc.)

Leverage FEMA GeoPlatform capabilities:

Viewer and reporting

Data archiving and access

Development and application environment

Upcoming Functional Enhancements:

1. Alignment of the Benefit Cost Analysis (BCA) andHazus-MH Flood Model Coastal depth-damagefunctions

2. Study region aggregation by CID community

3. Dasymetric general building stock exposuredistribution approach based on national land coverdata

Better Alignment with FIMA:

Better integrate NFIP and Risk MAP data directly with Hazus.

The Risk MAP National Flood Hazard Layer would be pulled directly intoHazus.

Incorporate NFIP approaches and rating methods

Provide reports integrated with NFIP impacts.

Improved reporting would be introduced to more directly feed hazardmitigation plans and better support benefit-cost analysis (BCAs).

Hazus MH BCAR Alignment (Task 2-Functional Enhancements)

Incorporate New Expert Panel Coastal A and V Zone Depth DamageFunctions (DDFs)

25

3

1. Benefit Cost Analysis (BCA) Alignment for Damage Functions

Hazus currently provides an extensive Depth Damage Function (DDF) librarythat was initially used for the BCA tools modernization effort several yearsago.

Our BCA program within the Mitigation Directorate at FEMA is criticallyimportant in ensuring that our mitigation investments and sound and costeffective.

This task will incorporate the latest additional DDFs from the BCA programrecently developed for coastal flooding.

This task enhances our alignment with FIMA programs and ensures Hazusloss projections can align with BCA benefit assessments.

Community Based Analysis (Task 2-Functional Enhancements)

Build and analyze study regions based on CID (Community Information Data)

What is CID?

•Based on 2010 Census jurisdiction boundaries

•Linked to NFIP’s Community Information System (CIS) that tracks community map adoption and participation

•Now incorporates tribes

•Aligns with other FEMA programs (NFIP, Mitigation Plan tracking)

26

4

2. Community-level Hazus-MH Study Region Aggregation

Currently, the Hazus-MH study region aggregation is by

• State

• County

• Census Tract

• Census Block (flood), and

• Watershed

The proposed Hazus-MH study region aggregation willadd

• State

• County

• Census Tract

• Census Block (flood), and

• Watershed

• Community

5

2. Community-level Hazus-MH Study Region Aggregation

Benefits:

Aligned with

FIMAprograms

NFIPcommunities

MitigationPlancompletionand trackingdatabase

It will allowforjurisdictionanalysis bycity

Non-incorporatedcounty land

tribal area

Dasymetric Methodology (Task 2-Functional Enhancements)

Simply removing undeveloped areas of census blocks for the loss estimation calculation

Benefits from 2010 inventory update and new Land Use Land Cover (LULC) release

Already completed a POC

Can be starting point for more inventory enhancements based on other LULC codes and point based approaches used by USACE and Hazus Tsunami Methodology

27

12

3. Dasymetric GBS exposure distribution approach

Use of NLCD Land Use Land Cover (LULC)

Suggested inclusion of NLCD 2011 LULC codes:

•21 – Developed, Open Space

•22 – Developed, Low Intensity

•23 – Developed, Medium Intensity

•24 – Developed, High Intensity

•81 – Pasture/Hay

•82 – Cultivated Crops

Suggested exclusion of NLCD 2011 LULC codes:

•The remaining 10 general LULCs, and

•The 4 LULCs for Alaska only

UPDATE: HAZUS Data & Training Collaboration

• NEW - HAZUS Sharepoint Site: has been set up to facilitate HAZUS data and training collaboration.

• Knowledge asset tool and an archive of Hazus knowledge.

• The goal of this effort is to increase its use and create a platform for deliberation and delivery of Hazus work products through the site.

This site will allow any registered user to (1) post a description of data/results they have to share with others,

(2) post a request of enhanced datasets or results that other’s might have already done,

(3) post information about an upcoming training, and

(4) post a request for training on a specific or general topic.

• To register, send Jenna McGee(jenna.mcgee@amec.com) your

(name, email, phone and organization) so she can build a list to generate initial accounts. Then we can deploy the Region III – NCR HUG group page!

8

Thursday September 25, 2014 at 10:00 AM EST

Use the same number and conference code:

Call 1-267-507-0240 and enter this conference code # 697620

**Request volunteer for next month’s presentation

• HAZUS / Risk Assessment project

• HAZUS “How to” presentation

Meeting notes and an updated presentation with audio will be sent out after this call

Adjourn

Next Call